CVE-2026-35572: ChurchCRM SSRF Vulnerability

CVE-2026-35572 Overview

ChurchCRM versions prior to 6.5.3 contain a Server-Side Request Forgery (SSRF) vulnerability [CWE-918]. The application accepts a crafted URL supplied in the HTTP Referer request header and uses it to initiate outbound HTTP/HTTPS requests to arbitrary hosts. Researchers confirmed the behavior using Out-of-Band Application Security Testing (OAST), where the server reached attacker-controlled domains. An authenticated attacker can abuse this flaw to probe internal networks, interact with internal services, or exfiltrate metadata from cloud environments. The maintainers fixed the issue in ChurchCRM 6.5.3.

Critical Impact

An authenticated attacker can force the ChurchCRM server to issue outbound HTTP/HTTPS requests to arbitrary internal or external hosts, exposing internal services and cloud metadata endpoints.

Affected Products

• ChurchCRM versions prior to 6.5.3
• Component: churchcrm:churchcrm
• CPE: cpe:2.3:a:churchcrm:churchcrm:*:*:*:*:*:*:*:*

Discovery Timeline

• 2026-04-07 - CVE-2026-35572 published to NVD
• 2026-04-10 - Last updated in NVD database

Technical Details for CVE-2026-35572

Vulnerability Analysis

The vulnerability resides in ChurchCRM's handling of the inbound Referer HTTP header. The application treats the header value as a trusted URL and passes it to a server-side HTTP client without validation against an allowlist or restriction of target hosts. As a result, the server initiates an outbound request to whatever destination the attacker chooses.

This class of flaw is tracked as Server-Side Request Forgery (SSRF) under [CWE-918]. SSRF in a web application allows an attacker to pivot through the vulnerable host. The server's network position, rather than the attacker's, dictates which resources can be reached. Common targets include internal HTTP services, administrative interfaces bound to localhost, and cloud instance metadata endpoints such as http://169.254.169.254/.

In ChurchCRM's deployment context, the server typically runs alongside a MySQL database and may be hosted in shared environments. An attacker can use the SSRF primitive to enumerate internal hosts, fingerprint services through response timing, or retrieve sensitive responses if any return content reaches the attacker.

Root Cause

The root cause is missing validation of attacker-controlled URL input before it is passed to the server's HTTP client. ChurchCRM accepts the Referer header value, parses it as a URL, and issues a request to the host specified, without enforcing host allowlists, scheme restrictions, or IP address filtering for private and link-local ranges.

Attack Vector

Exploitation requires network access to the ChurchCRM application and authenticated access at a privilege level the application accepts. The attacker sends a normal HTTP request to a ChurchCRM endpoint that triggers the vulnerable code path and sets a crafted Referer header pointing to the target host. The server then dispatches an HTTP or HTTPS request to that host. Confirmation via OAST in the original research shows the server reliably reaches arbitrary external domains.

No code example is published with the advisory. See the GitHub Security Advisory GHSA-44x3-28jv-mrwq for the upstream technical details.

Detection Methods for CVE-2026-35572

Indicators of Compromise

• Outbound HTTP/HTTPS connections from the ChurchCRM web server to unexpected internal IP ranges, including 127.0.0.1, RFC1918 addresses, or 169.254.169.254.
• Web server access logs showing inbound requests carrying unusual or externally hosted Referer header values that do not correspond to legitimate ChurchCRM pages.
• DNS queries from the ChurchCRM host to attacker-controlled or OAST-style domains such as *.oast.site, *.interact.sh, or *.burpcollaborator.net.

Detection Strategies

• Inspect HTTP request logs for Referer headers that contain IP literals, non-standard schemes, or hosts outside the application's own domain.
• Correlate inbound web requests with outbound network connections from the web server process within a short time window to surface SSRF-style behavior.
• Alert on any outbound request from the ChurchCRM server to cloud instance metadata endpoints, which has no legitimate use case for a church management application.

Monitoring Recommendations

• Forward web server, PHP, and host network telemetry to a centralized analytics platform and baseline normal outbound destinations for the ChurchCRM host.
• Enable egress logging at the network perimeter and on any cloud security group attached to the ChurchCRM instance, then review for connections to unexpected external domains.

How to Mitigate CVE-2026-35572

Immediate Actions Required

• Upgrade ChurchCRM to version 6.5.3 or later, which contains the official fix.
• Restrict outbound network access from the ChurchCRM server to only the destinations required for normal operation, blocking access to RFC1918 ranges and 169.254.169.254 where possible.
• Audit recent web server access logs for suspicious Referer header values and outbound connections that match the SSRF pattern.

Patch Information

The maintainers fixed CVE-2026-35572 in ChurchCRM 6.5.3. Patch details and the full advisory are available in the GitHub Security Advisory GHSA-44x3-28jv-mrwq. Administrators should review release notes before upgrading and validate the application in a non-production environment.

Workarounds

• If upgrading immediately is not possible, deploy a reverse proxy or Web Application Firewall (WAF) rule that strips or validates the Referer header on requests to ChurchCRM.
• Apply host-level egress filtering so the ChurchCRM process cannot initiate outbound connections to internal networks or cloud metadata services.
• Limit application access to trusted administrators only, since exploitation requires authenticated access per the CVSS vector (PR:H).